A patch welded onto a Michigan pipeline near Sterling Heights failed after leaking about 15 barrels of crude, and federal inspectors found the repair had no paperwork at all
A thin sheen crept across a storm drain near a busy intersection in Sterling Heights, Michigan.
A passerby was the one who noticed it.
Nobody at the pipeline company had seen it coming.
The estimated spill volume was 15 barrels, about 630 gallons, a small release that still triggered something the company had not faced before anywhere on its system.
No alarm had sounded in a control room hundreds of miles away.
The oil had already been seeping for some time before anyone with a badge or a sensor knew it existed.
What investigators found underground afterward would answer a bigger question: how does a buried repair fail for years without tripping a single sensor?
How does a buried steel patch fail without anyone knowing
A repair sleeve is exactly what it sounds like, a metal jacket slipped around a damaged section of pipe and welded shut to hold pressure the original wall could no longer contain on its own. Done correctly, it can last the life of the pipeline.
But done without adequate welding or inspection, it becomes a second seam, a new weak point hidden under soil where nobody looks again for decades. That is the mechanism investigators zeroed in on outside Detroit.
So why did the company’s own monitoring never catch it? A corroded seam under a sleeve releases oil slowly, not as a sudden pressure drop a control room algorithm is built to flag. PHMSA’s investigation traced the failure to a steel encasement sleeve welded onto the pipeline in 2010 to repair a defect, and the agency said the spill evaded Energy Transfer’s leak detection systems before a passerby noticed the oil.
What the pipeline itself looked like underground
The line running beneath the street is not new infrastructure. It is part of a 59 year old, 130 mile pipeline between Marysville, Michigan, and Toledo, Ohio, a route that has carried crude for nearly six decades.
That vintage matters. The pipeline used low frequency electric resistance welding, a manufacturing method federal regulators have long flagged as prone to integrity failures.
Even the volume involved tells a story about detection, not devastation. The estimated spill volume was 15 barrels, about 630 gallons, a small release that still triggered something the company had not faced before anywhere on its system.
How regulators found a nationwide problem inside one small spill
The corrective order followed fast. The agency issued the order three days after the Michigan spill was discovered, saying it was acting without giving Energy Transfer notice or an opportunity for a prior hearing because of the risk of “serious harm.”
Regulators do not usually skip that notice step. What pushed them to move so quickly was not the spill size but what the excavation turned up next to it. On Sept. 19, after excavating several locations along the pipeline, Energy Transfer confirmed the leak originated from a failed repair sleeve, a steel device used to fix damaged sections of pipe, that had been installed in 2010.
The company could not locate the original records for the Sterling Heights sleeve, nor could it verify the integrity of 11 other sleeves along the same Marysville to Toledo pipeline. Twelve welded patches, one documented history.
That gap, multiplied across a much bigger map, is what turned a local spill into a corporate wide directive. In an extraordinary move, due to a system wide risk of additional similar failures, PHMSA gave the company 90 days to submit a detailed inventory of past repairs made throughout its hazardous liquid pipeline network nationwide, as well as a plan to remediate vulnerabilities, a network that includes about 20,000 miles of crude oil, refined product and natural gas liquids pipelines.
Where this fits inside a pattern regulators are still tracing
This was not treated as an isolated weld failure. The same kind of encasement sleeve failure caused a jet fuel leak discovered last year on another Energy Transfer pipeline that contaminated the drinking water source of a residential neighborhood north of Philadelphia, the PHMSA order said.
Two separate states, two separate products, the same repair technique behind both. Regulators used blunt language to describe what that repetition implied, stating that failures under at least two sleeves with improper installation point to a likelihood of future failures.
Other operators have already treated this failure mode as reason enough to rebuild, not just repair. One competitor spent $50 million converting all Type A steel sleeves on its system to Type B sleeves, welded directly to the original pipe to contain pressure on their own.
What happens to the pipeline, and the record keeping, from here
For now the physical fix is narrow and already underway. The order directed Energy Transfer to shut down the affected pipeline segment, reduce operating pressure across the 130 mile stretch, and complete a series of inspections and repairs before the line can return to full service.
An agency spokesperson said the pipeline was restarted the following day at reduced pressure, slowing the flow of oil.
Much of the industry is only beginning to grapple with how fragile paper trails have become across decades old networks, a gap pushing operators toward digital recordkeeping that cannot simply vanish in a filing cabinet fire or a merger. An Inside Climate News investigation found the incident drew little political attention so far, a muted response compared with how loudly similar failures have echoed on aging networks elsewhere, including a Minnesota gas line rupture earlier this year.
The company has 90 days to tell regulators what it actually knows about its own welds. Until that inventory lands, every mile of sleeve repaired without a surviving record remains an open question buried under someone’s street, waiting for the next passerby to notice something nobody else did first.
Hugo is an engineer with strong technical expertise. Multilingual from an early age, his writing combines technical clarity with a strong interest in science and energy.
